Abstract: Elm sparse woodland steppe plays an important role in vegetation restoration and landscape protection in Horqin Sandy Land. In this paper, a two-factor and two-level field experiment was conducted to explore the effects of aboveground and belowground competition between grass and tree on the growth of e
lm seedlings in the Sandy Land. Five aspects were considered, i.e., seedling biomass, belowground biomass/aboveground biomass, stem height, ratio of root to stem, and leaf number. For the one-year-old elm seedlings, their biomass showed a trend of no competition > aboveground competition > full competition > belowground competition, belowground biomass / aboveground biomass showed a trend of belowground competition > full competition > no competition > aboveground competition, stem height showed a trend of aboveground competition > no competition > full competition > belowground competition, root / stem ratio showed a trend of belowground competition > full competition > no competition > aboveground competition, and leaf number showed a trend of aboveground competition > no competition > belowground competition > full competition. Belowground competition had significant effects on the growth of one-year-old elm seedlings, while aboveground competition did not have. Neither belowground competition nor aboveground competition had significant effects on the growth of two-year-old elm seedlings. It was suggested that in Horqin Sandy Land, grass affected the growth of elm seedlings mainly via belowground competition, but the belowground competition didn’t affect the resource allocation of elm seedlings. With the age increase of elm seedlings, the effects of grass competition on the growth of elm seedlings became weaker.

Key words: elm sparse woodland steppe, grass-tree competition, natural regeneration, Horqin Sandy Land